Resilient Quantized Consensus with Multi-Hop Communication

Abstract

In this paper, we study the problem of resilient quantized consensus where some of the agents may behave maliciously. The network consists of agents taking quantized/integer-valued states with asynchronous updates and time delays in the communication between agents. We propose a quantized weighted mean subsequence reduced (QW-MSR) algorithm where agents are capable to communicate with multi-hop neighbors. We provide necessary and sufficient conditions for our algorithm to achieve resilient quantized consensus for synchronous/asynchronous updates under the malicious attacks. Compared to existing methods in the literature, our method has tighter graph condition and, in particular, we establish that with multi-hop communication, the requirement for achieving resilient quantized consensus is less stringent. Numerical examples are given to verify the efficacy of the proposed algorithm.